EunJoung Byun

Characterizing and Classifying Desktop Grid

Desktop Grid has recently received the strong attraction for executing high throughput applications as CPU, storage and network capacities improve and become cheaper. Desktop Grid is different from Grid in many respects, but there is no general survey or taxonomy for desktop Grid. Therefore, we propose a new comprehensive taxonomy and survey of desktop Grid in order to characterize and categorize desktop Grid.

MJSA: Markov job scheduler based on availability in desktop grid computing environment

In a desktop grid computing environment, voluntary desktops (i.e., resource providers) are free to leave and join independently in the middle of execution. To develop a reliable desktop grid computing system, a scheduling scheme must consider the dynamic nature (i.e., volatility) of volunteers. Existing desktop grid computing systems, however, do not consider volatility in their scheduling procedures. As a result, job execution is often suspended, resulting in delayed completion time and degraded performance and reliability. To solve these limitations, we propose the Markov Job Scheduler based on Availability (MJSA) supporting three advanced scheduling schemes: OPTIMIST, PESSIMIST, and REALIST. These scheduling schemes are based on stochastic modeling of desktop availability. In the OPTIMIST scheme, in which time constraints are relaxed, the MJSA provides reliable resource selection at low cost. In the PESSIMIST scheme, where time constraints are rigid, the MJSA enables stable makespan in strictly time. Finally, in the REALIST scheme, where time constraints are only partially relaxed, the MJSA provides enhanced cost efficiency. In conclusion, the MJSA improves performance and reliability by adapting the appropriate scheduling scheme when selecting volunteers according to the needs of applications.

Scheduling Scheme based on Dedication Rate in Volunteer Computing Environment

A volunteer node can join and leave a volunteer computing system freely. However, existing volunteer computing systems suffer from interruptions of job execution, delays of execution time, and increases of total execution time, since they do not consider dynamic scheduling properties (i.e. volatilities), such as leave, join, and suspension. Therefore, dynamic execution properties of volunteer node should be considered in scheduling schemes, in order to design a stable and reliable volunteer computing system. This paper proposes a new scheduling scheme based on the dedication rate (DR), which reflects the dynamic properties of a volunteer. The scheduling scheme improves the completeness and reliability of execution, while also decreasing delay and total execution time. In addition, an implementation of the proposed scheduling scheme on top of Korea@Home is described, as well as performance evaluation

A Reliable Communication Protocol for Multiregion Mobile Agent Environments

A mobile agent system is regarded as an attractive technology when developing distributed applications. However, mobility makes it more difficult to trace agents. It is also more complex for agents to communicate with each other in a reliable manner. Therefore, a reliable communication protocol is necessary to control and monitor mobile agents and deliver messages between them. In this paper, a new reliable communication protocol (RCP) is proposed for a multiregion mobile agent computing environment. RCP is implemented on the ODDUGI mobile agent system. Analysis and evaluation show that RCP fulfills the following design goals: reliability, asynchrony, timeliness, location dependency, scalability, and communication cost.

ODDUGI: Ubiquitous Mobile Agent System

A mobile agent is regarded as an attractive technology when developing distributed applications in mobile and ubiquitous computing environments. In this paper, we present ODDUGI, a java-based ubiquitous mobile agent system. The ODDUGI mobile agent system provides fault tolerance, security, location management and message delivery mechanisms in a multi-region mobile agent computing environment. We describe the architecture, design concepts and main features of the ODDUGI. In addition, we present the One-Touch Campus Service application developed on the basis of ODDUGI in mobile and ubiquitous computing environments.

Agent-Based Autonomous Scheduling Mechanism Using Availability in Desktop Grid Systems

This paper addresses scheduling strategies to achieve high performance in desktop grid systems. The recent desktop grid systems can be characterized by decentralized control, large scale and extreme dynamism of their computation environment. In the environment of high throughput desktop grid systems, the volatility of volunteers, and the decentralized nature of desktop grid systems pose significant challenges. Our approach makes use of autonomous scheduling mechanisms on a computational overlay network to meet these challenges. However, current approaches to utilizing desktop resources require either centralized servers, or extensive knowledge of the underlying system, limiting scalability and performance. In this paper an agent-based autonomous scheduling (ABAS) mechanism on a computational overlay network is proposed to further improve performance through adaptive behavior of agents which have replication, migration, and checkpointing scheme. Performance evaluation demonstrates that the proposed scheduling mechanism improves throughput using Korea@Home

Autonomous scheduling through self-organizing computational overlay network in decentralized desktop grid systems

This paper presents construction of dependable computation environment and autonomous scheduling strategies in decentralized desktop grid systems. Desktop grid systems have the limits of performance because of characterized by heterogeneous, volatility, scalability, and extremely dynamic computation environment. In this paper, the Self-Organizing Computational Overlay Network (SOCON), through a coordinator and a volunteer based overlay network, is proposed to create a reliable decentralized computation environment. Moreover, the Autonomous First-Come First Serve (A-FCFS) scheduling on SOCON, which reflects dynamic volunteer characteristics, is proposed to further improve performance through the adaptive behavior of agents.

Probabilistic Observation Prediction Model based E4 Scheduling Mechanism in Peer to Peer Grid Computing

Peer to peer (P2P) grid computing is an Internet-based parallel computing paradigm to achieve tremendous computing power by voluntary peers. Since peers are not dedicated to a system, peers are free to leave and join autonomously during computation in this computing environment. A scheduling mechanism must cover the volatility of peers to support a reliable P2P grid computing system. Most existing P2P grid computing systems, however, suffer from blocked job execution and delayed makespan because they do not consider the volatile features of peers in the scheduling procedure. In order to improve system reliability, we propose a probabilistic observation prediction (POP) -based EA scheduling mechanism supporting three advanced scheduling policies: loose time constraint policy (LTCP), strict time constraint policy (STCP), and hybrid time constraint policy (HTCP). LTCP is for general use, STCP aims at minimizing makespan, and HTCP attempts to improve efficiency, and performance-to-cost ratio. All policies are based on stochastic modeling of peer state transition while endeavoring to manage the volatile features of peers

Advanced stochastic host state modeling to reliable computation in global computing environment

To design a stable global computing environment supporting reliable job execution and covering unanticipated state changes of hosts, the dynamic characteristics (i.e. volatilities) of hosts should be considered. Since a host is not dedicated to the system, voluntary hosts are free to leave and join autonomously in the middle of execution. As current systems do not relate volatility to a scheduling procedure, global computing systems suffer from performance degradation, reliability loss, job interruption, and execution time delays. For dependable computation, we propose Advanced Stochastic Host State Modeling (ASHSM), which is based on Markov model relating to execution availability quantifying duration and regularity of execution patterns of each host. Through the model, the system predicts desktop activities and allocates jobs according to the host features. Furthermore ASHSM alleviates unreliability due to unstable resource provision and job suspension during execution

MTRMS: mutual-trust-based reputation management system in community-based peer-to-peer environment

Light-scatter smoke detector, depending from the ceiling of a room, has a base unit with a battery to power an infra-red light source and infra-red light sensor, both of which are located in an upper element. A lower element has a surface corresponding to the exterior of a cone, and forming the base of a chamber. A surface makes, with a frusto-conical surface of the upper element, an annular wedge-shaped recess to chamber, this recess facing both source and sensor. The surfaces have a matt-black coating to promote energy absorption, so that the recess inhibits any light entering it from ever re-emerging. The stream of air flowing through the detector is monitored in the central chamber for the presence of smoke particles, by the sensor watching for light which originated from the source being scattered by smoke particles in the chamber and arriving at the sensor.